Twenty-five years ago, when the case of Bakke v. University of California arrived before the U.S. Supreme Court, it was widely anticipated that the Justices would at last resolve an issue that had been bedeviling the country for years: the permissibility of preference by race in university admissions. It did not happen. To the contrary, the internal divisions of the Court at that time, as reflected in six tangled opinions, left the matter in a more muddled condition than it already was. True, Allan Bakke, the white applicant who had been turned down by the University of California in favor of less qualified minority candidates, won his suit; naked racial preference was thrown out. But what other sorts of racial and ethnic preferences might be permitted was left quite uncertain.
The chief muddler in 1978 was Justice Lewis Powell, a decent man and an honorable judge who found racial discrimination appalling and unconstitutional and yet also felt that he had to permit some wiggle room for college admissions officers to attend to race under some circumstances. In his long and convoluted opinion, notorious for the confusion to which it subsequently gave rise, Powell held that it would be reasonable for a university to take into consideration the race of particular applicants for the sake of achieving intellectual “diversity” in the student body. To treat people in general differently because of their color, Powell said, was plainly a violation of the equal-protection clause of the Fourteenth Amendment. But to allow the race of individual applicants to weigh in their favor for the sake of diversity did not amount to such a violation.
No other Justice joined Powell in his confused and rather fanciful homage to the concept of diversity. And yet his principle took root. This was in part because, in a Court divided between two parties of four, his had been the deciding voice. For the universities, the problem with Bakke was that it unambiguously rejected preferences for the sake of remedying past injustices committed against racial minorities—precisely the defense that, until then, many universities had been relying upon. If they were determined to go on giving preference, as for the most part they were, they would henceforth have to lean upon the weak reed of Powell’s speculations concerning diversity. And this, for the next quarter-century, they proceeded to do.
But would the “diversity defense” withstand renewed constitutional challenge? That central question was presented to the Supreme Court in two cases involving the University of Michigan and finally decided this past June: Gratz v. Bollinger, concerning the admissions practices of Michigan’s undergraduate college, and Grutter v. Bollinger, concerning the admissions practices of its law school. (“Bollinger” is Lee Bollinger, formerly the president of the university.) Though the two cases differed significantly in their particulars, in each case the university justified its practice of using racial preferences on the grounds not of remediation but of diversity.
One would have thought—I certainly thought—that the university would have an extremely tough time of it. For any state to treat people differently by race is an odious practice, presumptively unconstitutional and hence subject to the rigorous standard of “strict scrutiny” applied to any putative exceptions to the rule. That standard has two prongs. To win the day, the university would need five of the nine Justices to agree both that racial diversity was a “compelling” need of the state of Michigan and that the system of preference it was using had been “narrowly tailored” to meet that compelling need.
How could it do that? It seemed eminently plain that the state’s need for racial diversity in its university, if it had any such need at all, was not compelling under any ordinary meaning of that term. Moreover, the two Michigan systems giving preference, so far from being narrowly tailored, were (as one federal judge had earlier put it) more like a chain saw than a sewing machine in their mode of operation. Success for the University of Michigan thus seemed very unlikely indeed.
So confident was I that five Supreme Court votes could not be gathered to support a view plainly concocted only to pass muster under Powell’s 1978 opinion that two years ago I came as close as the editors of COMMENTARY would permit me to predicting certain defeat for the university (“Race Preferences & the Universities—A Final Reckoning?,” September 2001). As I saw it, no judge, taking seriously the standard of strict scrutiny, could possibly approve diversity as a state need compelling enough to justify deliberate racial discrimination. Right and left, I wagered steak dinners with every soft-headed supporter of the diversity principle who would allow his political passions to overrule his good sense.
I am buying many steaks these days.
In Gratz v. Bollinger, the Supreme Court of the United States did hold that the numerical admissions system used by Michigan’s undergraduate college, in which a given number of points was awarded to all applicants in certain ethnic categories, violated the equal-protection clause of the Fourteenth Amendment as well as the Civil Rights Act of 1964. But on the same day, in Grutter v. Bollinger, the Court held that “the educational benefits that flow from a diverse student body” were indeed, in the context of higher education, a compelling state interest. Moreover, the particular form of deliberate racial discrimination practiced by the law school of the University of Michigan was found to be consistent with the constitutional guarantee of equal protection of the laws.
The diversity principle, even if only in one context, and with heavy restrictions, has thus been embedded in law; Powell’s weak reed has become a mighty limb. In the meantime, in the tension between these two latest decisions, what was muddy in Bakke has become muddier still.
Much damage has been inflicted on the standard of strict scrutiny itself. As the four dissenting opinions in Grutter (the law-school decision) make vividly, even bitterly, clear, the need of the state of Michigan for diversity of skin colors in its university admissions was hardly “compelling.” Even the maintenance of a law school by the state of Michigan, as Justice Clarence Thomas pointed out, although surely a good thing, is not a compelling need; many states thrive without one. Nor, for that matter, has the university’s program delivered on its promise: exhaustive research has shown that its purported benefits (i.e., the promotion of tolerance and understanding for the views of “diverse” others) have been in scant evidence. No genuinely “compelling” educational need is being served.
As for the second prong of the standard of strict scrutiny—the requirement that, even if some compelling state need has been identified, the use of racial classifications must be narrowly tailored to the fulfillment of that need—the race preferences given by the law school, no less than the rejected admissions system of Michigan’s undergraduate college, likewise failed to satisfy it. Demonstrating this failure formed the nub of Chief Justice Rehnquist’s dissenting opinion in Grutter, in which he was joined by Justices Clarence Thomas, Antonin Scalia, and Anthony Kennedy. It has not been widely understood.
According to the University of Michigan, the educational benefits of diversity were said to flow from the creation of a “critical mass” of minority representatives in the student body: a number sufficiently large that the members of a given minority in a given class would not feel themselves “isolated” in that class. The racial instrument used in law-school admissions was, the university held, narrowly tailored to this end. Unlike the practice in the undergraduate college, the law school, in evaluating applicants, claimed not to rely on some fixed numerical value mechanically awarded to every member of certain ethnic groups. Instead, the critical masses needed for the three designated minorities—African-Americans, Native Americans, and Hispanics—were said to have been artfully assembled through the use of highly sensitive reviews of each individual applicant.
Indeed, in the earlier, district-court trial of Grutter, as in the university’s written arguments for the Supreme Court, the law school had gone to great lengths to avoid any mention whatsoever of numbers or percentages. And for good reason: any open confession of this kind would have exposed the school to the same condemnation received by the undergraduate college in Gratz. But the entire law-school system was a deception. The dissenting opinion of the Chief Justice proved this.
In any given class of students, a “critical mass,” whatever it is, cannot be greatly different in size for African-Americans from what it is for other minorities; nor can it differ greatly in one year from what it was three years earlier, at least if the class itself has not changed in size. In fact, however, the numbers of the several minorities admitted to the Michigan law school in order to form their respective “critical masses” differed very greatly from minority to minority and, for each minority, from year to year. Calling attention to this starkly revealing feature of the school’s admissions over a period of many years, and inserting numerical tables into his opinion to render the matter incontrovertible, Justice Rehnquist showed that the number of those admitted in each minority closely tracked the number of applications by members of that minority.
The figures themselves are worth examining. In recent years, the Michigan law school has offered admission each year, on average, to sixteen Native Americans, 51 Hispanics, and 100 African-Americans. If what had really been sought was a critical mass of each minority—that is, a number sufficiently large to ward off feelings of isolation within the class—and if the yield of 50 offers was enough to achieve that in the case of Hispanics, it cannot be the case (as the Chief Justice pointed out) that the yield of 100 offers was needed among some other minority, or that the yield of sixteen offers would be sufficient in the case of still another.
How did the numbers themselves come about? Rehnquist described the process. When, in 1995, 9.7 percent of the applicant pool was African-American, 9.4 percent of those admitted were African-Americans. Five years later, when 7.5 percent of the applicant pool was African-American, 7.3 percent of those admitted were African-Americans. A similar pattern was manifest throughout.
This “tight correlation between the percentage of applicants and admittees of a given race,” Rehnquist wrote, devastatingly, could
only be the result of very careful race-based planning. . . . We are bound to conclude that the law school has managed its admission program, not to achieve a “critical mass,” but to extend offers of admission to members of selected minority groups in proportion to their statistical representation in the applicant pool. [emphasis added]
Despite the law school’s repeated protestations to the contrary, the statistics demonstrate that it was seeking some proportional representation of minorities in its entering classes. Its real objective was not critical mass—this, in Rehnquist’s words, was “simply a sham”—but racial balance. But it could not have proclaimed this obvious truth, contending forthrightly that such proportionality would serve as its achievement of “diversity,” for the simple reason that the admission of persons simply to achieve certain percentages of various ethnic groups is “patently unconstitutional.” Justice Powell had made that point crisply in Bakke decades ago, and the same proposition was affirmed by the majority in Grutter without reservation. Instead, the law school engaged in a sham.
The Grutter majority accepted the sham whole hog. Baldly asserting that student-body ethnic diversity was a compelling state need, it found this end narrowly served by a process in which, allegedly, ethnicity was considered as no more than a “plus factor” in the files of “particular applicants,” all the attributes of each applicant being “holistically” appraised and “all pertinent elements of diversity considered in light of the particular qualifications of each applicant” in the quest for a critical mass. (The language is that of Justice Sandra Day O’Connor in her majority opinion.) On this reading, the fact that, of two law-school applicants with identical academic credentials, a black applicant’s chances of admission were in fact hundreds of times greater than those of a white applicant must be regarded as no more than a coincidence.
If this is strict scrutiny, the term is without meaning.
Thanks to the Court’s decision, still worse is now to come. The law-school system, a sham in reality, has been elevated to the status of a model—and not only for law schools. As we have seen, Michigan’s use of race in undergraduate admissions, which did not pretend to be “individualized” or “holistic,” was on that account found to be flatly unconstitutional. But henceforth, Michigan and many other universities will formulate their undergraduate preferential schemes in phrases echoing the language of the law-school program. Beginning now, “individualized review,” “holistic,” “critical mass,” “plus factor,” “a particular applicant’s file,” and the like will appear ubiquitously and talismanically in the description of admission systems from coast to coast.
Of course, it will be far easier to profess such highly individualized review systems than to realize them. And so a second-level sham will be explicitly invited: a fraud imitating a fraud.
Consider: the law school at Michigan enrolls some 350 new students each year. Of the several thousand who apply, a good number are speedily disqualified, with many of the remaining applicants interviewed in person and the complete file of every admitted applicant examined by a single person, the assistant dean for admissions. Even though, at the Michigan law school, the real goal has been racial balancing, the requisite process of “individualization,” as approved by the Court, is conceivably doable, if with some strain.
The undergraduate college at Michigan is a rather different affair. It receives more than 25,000 applications for admission each year. Picture a gymnasium in which those fat application files are stacked in piles six feet high; there will be some 350 of these piles, or more, pretty nearly stuffing the gymnasium to the gills. Now imagine that each application is to be evaluated comparatively, with race and many other factors given varying and appropriate weights in the assessment of each candidate. Remember, no numerical value for ethnicity is to be assigned, no quantitative system applied.
How, in the name of reason, is the comparison of these 25,000 applicants to be carried out? Even for an army of admissions officers, the exercise would be hopeless. It is utterly impossible for the University of Michigan—not to speak of universities in Minnesota and Ohio and other states where undergraduate colleges are substantially larger still—to review all the particular qualifications of each of tens of thousands of applicants, weighing race as but one factor, without using some numerical calculus. Any future claim to that effect is guaranteed to be a deception.
In its written argument defending the mechanical award of points for race in undergraduate admissions, the University of Michigan granted candidly that “the volume of applications and the presentation of applicant information make it impractical for [the undergraduate college] to use the . . . admissions system” of the law school. Of course. That is why the university argued, in effect, that since the racial results it sought could only be achieved using a system of numerical weights, such a system must be permissible, for there is no other way to achieve the approved aim of diversity.
No! responded the Court in Gratz. The use of race is permitted in some ways, but it is not urged, and you are certainly not entitled to do whatever you think is required. The Court’s strictures were not to be bypassed: a university may not “employ whatever means it desires to achieve the stated goal of diversity without regard to the limits imposed by our strict-scrutiny analysis.” Nor did “the fact that the implementation of a program providing individualized consideration might present administrative challenges . . . render constitutional an otherwise problematic system.” So, under Gratz, the university has been forbidden to do what it asserts it must do in order to achieve the racial objective it asserts it must pursue and which, under Grutter, has now been found “compelling.” Here, in the pull of the two decisions against each other, is indeed a recipe for still more pervasive obfuscation and more shameful hypocrisy.
Justice Ruth Ginsburg, who dissented from the majority in Gratz, saw this clearly. Writing in support of the now-unlawful point system, she frankly acknowledged that universities have already been deceitful and sly in this arena, resorting to “camouflage” by encouraging minority applicants and their supporters to convey their ethnic identification deviously and backhandedly in personal essays and letters of recommendation. Justice Ginsburg then concluded: “If honesty is the best policy, surely Michigan’s accurately described, fully disclosed college affirmative-action program is preferable to achieving similar numbers through winks, nods, and disguises.” Narrow tailoring need not be faked; instead, in Ginsburg’s view, it could simply be ignored. A non-individualized program, assigning a fixed number of points for skin color, was the answer. Unless we permitted it, the result would be widespread cheating.
She was certainly right about the cheating. But cheating is already endemic, and is now bound to spread further. For those bent on racial preference, the “winks, nods, and disguises” decried by Ginsburg have now, thanks to her and her likeminded colleagues in Grutter, been made a practical necessity.
In a footnote to the opinion striking down the undergraduate point system, Chief Justice Rehnquist called Ginsburg’s observations “remarkable,” and answered them sharply:
First, they suggest that universities—to whose academic judgment we are told in Grutter we should defer—will pursue their affirmative-action programs whether or not they violate the United States Constitution. Second, they recommend that these violations should be dealt with, not by requiring the universities to obey the Constitution, but by changing the Constitution so that it conforms to the conduct of the universities.
This footnote, destined to become famous, goes to the heart of the two cases. Conduct that is plainly wrong and ugly if confronted openly has been condoned by five members of the Supreme Court if carefully hidden and deliberately misdescribed. The unconscionable deceptions of years past are now enshrined and soon to be compounded. Duplicity is encouraged to run amok.
To be sure, there are some signs of judicial distress. Even Justice O’Connor, who in Grutter found race preference tolerable, reluctantly acknowledged that “there are serious problems of justice connected with the idea of preference itself.” These are precisely the problems that will saddle universities, and American society, for years to come.
How many years? The Court accepted the assurances of the University of Michigan that the racial discrimination it now practices must some day end. The firmly expressed expectation of the Court is that this will happen within 25 years (a piety to which Justice Thomas retorted that the principle of equality ought not have to wait a quarter of a century to be vindicated). In the meantime, what lies ahead is the agony of a long chain of public disputes. For one thing, there will be no end of quarreling over the systems of preference being given. Although the universities will claim the protection of the words used in Grutter, it will be very hard to hide the reality of their practices, and these will be subjected to continuing adverse scrutiny. We may thus be reasonably certain that Michigan and those of its sister institutions likewise relying upon allegedly nonquantified diversity as the justification for preferences will be back in court again and again.
But the controversy will also move from the courtroom to the ballot box. If the Supreme Court has found that, in the interest of diversity, race preference may be given, it remains for the people of the several states to decide for themselves whether, in their state, race preference is to be forbidden. In Michigan, for example, every effort will be made by the time of the presidential election of 2004 to place on the ballot a Michigan Civil Rights Initiative—an equivalent of California’s Proposition 209. The operative sentence in that proposition, now incorporated in the California constitution, is nearly identical to a critical passage of the Civil Rights Act of 1964, with the addition of five words that appear here in emphasis. It reads:
The state shall not discriminate against, or grant preferential treatment to, any individual or group on the basis of race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting.
Once the matter is on the ballot, it will also become more difficult for legislators and political candidates to dodge this controversy as they have so often done in the past. Will they urge their constituents to vote against a proposition forbidding race preference? If so, must we not conclude that they support race preference?
The decision of the U.S. Supreme Court in Grutter v. Bollinger is disheartening in the extreme. But the governing rule in this matter will come ultimately from the citizenry, and we must trust that the large majority of Americans, as reported in survey after survey and confirmed in election after election, continues to find racism of every sort disgusting. I was wrong about the outcome of the battle in court; now the war must move to other fronts.